Algorithms to AI in Video Analytics (S1:E40)
November 3, 2022
Our new episode is all about intelligent video analytics, technology that processes digital video signals using a special algorithm to perform a security-related function. Identiv’s Mike Taylor, VP Global Sales, and Mark Allen, GM Physical Access, join us to talk about the latest security trends in video content analysis, from facial recognition and asset tracking to AI learning.
Full Transcript
Speaker 1:
You're listening to Humans in Tech. Our podcast explores today's most transformative technology and the trends of tomorrow, bringing together the brightest minds in and outside of our industry. We unpack what's new in physical access, identity verification, cyber security, and IoT ecosystems. We reach beyond the physical world, discuss our digital transformation as a species, and dive into the emerging phygital [00:00:30] experience. Join us on our journey as we discover just how connected the future will be and how we will fit into that picture. Your host is Leigh Dow, VP of Global Marketing at Identiv.
Leigh Dow:
Thanks for tuning in. Today we're talking to Mike Taylor, Vice President of Global Sales, and Mark Allen, General Manager of Physical Access, both leaders on the Premises Team here at Identiv. Thank you for being here today.
Mike Taylor:
Thanks for having us.
Mark Allen:
Good to be here.
Leigh Dow:
First off, maybe starting with Mike, fill us in on your roles here at Identiv [00:01:00] and who you manage on a day-to-day basis.
Mike Taylor:
Thanks, Leigh. So, as Vice President of Global Sales, I am one half of the table. Mark Allen runs the physical product side of the business, and he'll explain to you what that is, and I have everything that is sales. And what that means is I've got the field sales team that's sell on the commercial side, I have the federal sales team that sell into fed government and specific agencies. I'm fortunate enough to have an amazing team of sales engineers [00:01:30] that support the sales team. I've got an inside sales team and a sales operations organization that's amazing. I've also been able to just create a team that goes after our card business specifically, as well as some that sell third party items for us. So, very fortunate. I've got a team of about 52, 26 of those who are direct into me, and we are charged with just making the sale. The tip of the spear, if you will.
Leigh Dow:
Nice. How about you, Mark?
Mark Allen:
Yeah, thanks, Leigh. Yeah, I'm basically [00:02:00] Mike's personal assistant, as he mentioned. I'm the other side of the equation here, so anything the sales guys need, we're here to help. We have a product management team that listens to all of the voice of customer issues and turns those into great products that Mike can sell. I have the tech support team, so that we can support our channel and our customers directly. We have other supporting teams like training and tech [00:02:30] services as well. So, all of the roles that are there to help Mike get out there and sell a great product.
Leigh Dow:
Excellent. Mike, today we're discussing the trend of video analytics, also known as video content analysis and intelligent video analytics, technology that processes a digital video signal using a special algorithm to perform a security related function. What are the most common types of video analytics?
Mike Taylor:
Well, Leigh, there are several. Most of the time we're talking about fixed analytics, but there's [00:03:00] also algorithm analytics. There is AI learning, which is kind of the new buzzword, most people are pushing into the AI learning algorithms, as well as facial recognition systems and asset tracking as well. So there's a wide swatch of video analytics out there that we work with and we employ when we're deploying our systems.
Leigh Dow:
And Mark, how does that impact access control, particularly when you're talking about end-to-end access control?
Mark Allen:
Great question there, Leigh. This is where [00:03:30] we can actually multiply the video analytics value in the marketplace. A lot of people watch video and they can react to video, they can see an incident going on and they can do something about it. Also, video analytics can actually be proactive and look at things. You can program a video analytic to say, "Let me see if somebody's moved an object or they're going the wrong way down an aisle." But then the other side of the equation is, "Well, what does physical access do in response to that analytic [00:04:00] going off?" And that's where a complete solution really works well for people. When you can feed into your access control solution that, "Hey, somebody is outside of your building and they're doing something that looks kind of nefarious," you can then react with your access control system and maybe change the threat levels. Maybe go from threat level green to orange, maybe that's interfaced with an intercom system that lets everybody know that there may be something going on the left side of the building.
If you see something really [00:04:30] bad, there's been some incidents in the last year or so, the one that comes to mind mostly is the January 6th incident that happened at the Capitol Building. If video analytics was sitting there watching crowds form in an area where they really shouldn't be, or even as Mike mention, some of the more advanced video analytics that are coming on now where you can actually listen to people talking through the video cameras, and you heard maybe some of those people saying, "Hey, let's get through that window, or let's go through that [00:05:00] door," that could be an analytic driver that goes into access control that forces a lockdown, and maybe that complete automated system then could be responsive more quickly than what happened, where we saw real people had to then make analytic decisions.
Leigh Dow:
Is it at the point yet where it picks up on the word, like a trigger word or something like that?
Mark Allen:
Yeah, it is. We've seen analytics out there now that's not only using the cameras for looking but also using [00:05:30] the cameras for the voice too. So yeah, you can actually pick up on people speaking and turn that into an analytic alarm.
Leigh Dow:
Oh, that's really interesting. So, video analytics are more commonly used, like most people when they think about video analytics, if they know anything about it, they're thinking about how it's used on social media platforms like TikTok and YouTube. But what are some of the most recent improvements in video analytics?
Mike Taylor:
Well, I think everyone knows I'm a big fan of TikTok. I'm [00:06:00] posting [inaudible 00:06:01].
Leigh Dow:
Posting constantly.
Mike Taylor:
Exactly. I think the really key part is you've got to look at the intent of the analytics. There's a lot that goes on behind the scenes with TikTok, who owns it, where's that product based, and what are they doing with that data? Whereas we're always looking from our side of the businesses, how do we take these analytics and plug them into a full end-to-end solution? Because most of our customers come to us and they need to do more with less. Everyone's budget's [00:06:30] cut, everyone has fewer people, so we're looking to leverage it and be able to take these tools, essentially maximize our force, actually double or triple our human force by leveraging these technologies.
I think the really key part for us is being able to work with different analytics, work with different manufacturers. Because our VMS is an open platform, we really have the ability to let the customer select exactly what is best for them, what's the most [00:07:00] impactful solution that they're going to be able to get? We have a deep integration with that platform. And then again, we can take all of that data and feed it through the entire system, as Mark was saying, on the end-to-end solution, tying it into other systems as well. So, it's a lot of the underlying analytic technologies are the same, but it's really the use case for those technologies that are different.
Leigh Dow:
Interesting. Just fun fact, I actually don't post a lot on TikTok, but I could go down a TikTok rabbit hole and spend hours on [00:07:30] there searching and stuff. I get some of my best recipes from there. So, Mark, you and I have talked a lot in the past about the advantages of implementing video analytics and what they can do and that, yes, you have the security aspects of it, but it can also be applied to just the way you manage a building or people in the building, or crowds or things like that. So what are some of the advantages of implementing video analytics?
Mark Allen:
Yeah, just to expand on [00:08:00] some of those ideas. As resolutions have gotten better and pixels are more specific to what's in a picture, what happens then is the engineers that are behind the scenes looking at those pixels can actually make better decisions on the things that are happening inside of the picture. So, when you had a not so good resolution, you could really just watch blobs and guess whether that blob is a person, or whether that blob's a child or [00:08:30] an adult. As the whole industry advances, that technology really can be a lot more specific. You can start to see people carrying things, you can tell the difference between types of people. You can profile people. I know I probably shouldn't be saying any of this stuff, but that's what these systems can do. They can check whether you've got young people in the area, old people in the area, whether they've got glasses, all of those things.
As you get those better [00:09:00] resolutions, you can be a lot more specific with your analytic. You can look for specific types of people doing specific types of things, and then those become alarms and events that you can make access control actions on. So it's really getting to the point where we can do a lot of things with video, and as you integrate with the complete solution, then it becomes both the input and the output. The video is the input to the system, and then the access control [00:09:30] can be the output, where it actions on the events that are taking place.
Leigh Dow:
So it's that intelligence component of it, that's-
Mark Allen:
Absolutely.
Leigh Dow:
Yeah, that's important. So, Mike, just with respect to that, when you talk to customers, you meet with customers, what are the common application areas and industries that are really looking for video analysis?
Mike Taylor:
I really think as we get out and meet with customers, the more we can explain [00:10:00] to them the benefits and the tools that are at their beck and call, if you will, it really crosses over just about every type of industry. So it isn't one vertical market more so than the other. In fact, what we're seeing is, for the first time, we call generally on facilities or security people, and as the IT group has come into the fold, now we're able to give these people not just the tools that they need, but actually a lot of times we can give them what they [00:10:30] don't even know they need. And what I mean by that is there are times where they'll say, "Hey, we want an analytic that tells us if someone falls on the floor, we want to catch it. We want to know if it was a real fall, if it was a fake fall."
Well, if you're looking down an aisle at a store, that's great, but we're also, at that point we're recording everything that they're doing. So why don't we also take some of that same data using some background AI and start giving you an idea of which areas they spend more time? "Hey, in this area they're spending much more [00:11:00] time, what product do we have in that area that is of higher value than others?" And so now we're able to give the customer not only what they knew they want, which is help secure the business, maybe drive down their risk of loss and potentially mitigate risk as a whole, but now we can actually help jumpstart them on the other side of their business and become a tool that they can use to grow their business on top of just worrying about shrink or loss.
Leigh Dow:
Oh, that's really interesting. So like in a retail environment, [00:11:30] per se, somebody could know where the traffic is and that's where we should put certain products?
Mike Taylor:
Yeah, we're doing everything today from telling them which endcap has a longer hold rate. So, "Oh, this endcap, they like it. What are we doing there?" To, "Hey, listen, we can also tell you what time of day you have longer lines." Certainly we have a specific bank that wants to know when a certain line gets a certain length and what the most peak time happens to be, so that they can adequately stock it with tellers [00:12:00] or with cashiers. So it is really becoming a business tool across the board to help them better manage their business, and it's just because all of the data that we're already gathering, let's just use that data in a different way than we have in the past.
Leigh Dow:
So really, when you're talking to end users and customers now, like you said, it's not just the security people, it's also the IT people, probably operations people, facilities people?
Mike Taylor:
Across the board. You're absolutely right. I mean, we've gone [00:12:30] from the dusty operations office, maybe I'm setting my laptop on a file cabinet, to we spend much more time in the boardroom today with CIOs, CTOs, CEOs that want to help better the bottom line or mitigate risk. So, my meetings are usually 10 to 12 people, there's representatives from multiple different departments there, because this is not a security system, this is a business system, and it impacts their business across the board.
Leigh Dow:
[00:13:00] Interesting. Really part of that digital transformation story that businesses are trying to get to today?
Mike Taylor:
Completely, absolutely.
Leigh Dow:
So, Mark, what should businesses consider when implementing intelligent video analytics solutions for their organizations?
Mark Allen:
It's very dependent on the business itself. Obviously, if you're deploying access control solutions and video solutions to do security, then obviously it's very different than when you're deploying it for [00:13:30] bettering the business. But they really should just consider getting the best equipment they can, getting something that actually has a future in it. Look for product. When you're talking to the salespeople, they're not saying, "Okay, well, this is our last release." Obviously they wouldn't say that, but look for companies that are looking to extend their analytic set of tools going forward so that you can expand your solution [00:14:00] going forward. You really don't want to settle for something that's not going to help you move your company forward. What you'll find also is that as a company that's deploying video solutions, you'll put it in and then immediately you'll go on, "Oh, there's 10 more things I want to use this for." So really be ready for that surge of functionality that you're going to want to add to that video solution.
Leigh Dow:
And Mike-
Mike Taylor:
Yeah, Mark, I think, hold on. It just hit me, Leigh, I apologize. [00:14:30] But the question is what should you consider? And Mark really nailed it, and that is you've got to consider the network, you've got to consider the cameras, you've got to consider the storage, and again, you need to do it as such that you give yourself the ability to expand. I think that's the biggest thing, right? Pick the right technology, pick an open platform, pick something that you can integrate with multiple different systems, because what's available today, there'll be brand new systems available tomorrow and you don't want to be locked in a dead-end solution, is I think what Mark was getting at, and I think [00:15:00] he was spot on.
Leigh Dow:
Awesome. That's what I was going to ask you, if you had anything to add to that.
Mike Taylor:
I always have something to add, Leigh.
Leigh Dow:
I know, I know, I know. So, Identiv's award-winning Velocity Vision is a really powerful end-to-end solution and it's been recognized for enabling real-time visual intelligence to detect threats, to mitigate loss, and improve overall security operations. Do you want to fill us in on any upcoming products adding to the success of Velocity Vision?
Mark Allen:
Yeah, maybe I should take that one. Velocity Vision, [00:15:30] we started last year. It's been a really great add-on for us as a company. As you can probably tell, both Mike and I are both very excited about the things that we can do with Velocity Vision today, and to Mike's point, we have spent a lot of time making a complete solution. You don't just come to Identiv's for a VMS system, you come to Identiv for the complete solution. We can put it on a platform that is highly available, we can put it on a platform that easily [00:16:00] can be moved around and expanded and managed and integrated, which then goes into that complete solution. We do have a new release of Velocity Vision coming out at the end of the year with some really great feature set coming out on it, and each year we'll be releasing new Velocity Vision solutions with lots of extra functionality.
Leigh Dow:
Thanks for joining us, and we appreciate you taking the time out of your day to participate in our Humans in Tech [00:16:30] Podcast.
Mike Taylor:
Thanks for having us.
Mark Allen:
Thank you, guys.
Leigh Dow:
Yeah, for sure. If you enjoyed this podcast, please like and subscribe. We drop a new episode every Thursday.
Speaker 1:
The problem isn't security, it's awareness. Velocity Vision is the future of visual surveillance, an intelligent video management solution that delivers real-time situational awareness in an open-security platform. Integrate with your existing systems, verify your environment in one pane of glass, and increase the efficiency [00:17:00] of your security operation. Get full control of your environment when and where you need it. Learn more at identiv.com.
Get access control anywhere, anytime for less money out of pocket. Highly-secure Freedom Cloud is a cloud-based access control as a service offered through a cost-effective subscription model, allowing users to control, manage, and maintain their physical access control systems via Freedom's intuitive, always up to date [00:17:30] browser-based web administration. Learn more at identiv.com.
Physical security, identity verification, the IoT. The hyperconnectivity of our lives will only grow more pervasive. As technology becomes more automated and experiences more augmented, it's up to us to preserve our humanity and use new tools and trends for good. The only question is, are we up for the challenge?